Our laboratory investigated the earliest phases of hypertension in order to separate primary from secondary (pressure-induced) phenomena. Repeatedly, we found abnormalities of the autonomic control of hemodynamics combined with characteristic personality traits in patients with borderline hypertension. Due to invasive methodology, progress was slow and it took us 18 years to complete measurements in 220 BH and 190 control subjects. The possible preselection of volunteers for invasive studies could have affected the general applicability of our results. Furthermore, not all borderlines will develop established hypertension, and it is not known which of the abnormalities found in the laboratory are relevant to future hypertension. Recently the situation has changed. a) New, reliable noninvasive hemodynamic methods were cross-validated in our laboratory. b) We have new methods for assessing target organ damage and the patients' average BP levels. This permits stratifying BH by "severity" and thus by degree of likelihood to develop "true" hypertension. c) Tests to evaluate sympathetic function are simplified and applicable to large studies. d) Therefore, it is now possible to mount a field study in the village of Tecumseh, Michigan, where extensive records on BP in families are available. The goal is to organize a population-based study of physiologic and behavioral precursors of hypertension. We wish to know a) the prevalence of the hyperadrengergic state among unselected subjects with borderline hypertension, b) are personality traits (submissiveness, anger) found in the laboratory also descriptive of and specific for free-living patients with borderline hypertension, c) are the hyperadrenergic state and a specific personality pattern associated, and how do they relate to risk of hypertension? The second aim of this proposal is to select subjects with highest risk for hypertension and study them in detail. They will be reexamined after three years in order to learn about the stability of psychophysiologic interactions in borderline hypertension, and to draw inferences about pathophysiology of hypertension.